caa a22 4500 605541213 CHVBK 20210128100915.0 cr unu---uuuuu 210128e20150501xx s 000 0 eng 10.1007/s00371-014-0998-2 doi (NATIONALLICENCE)springer-10.1007/s00371-014-0998-2 Automatic blur-kernel-size estimation for motion deblurring [Elektronische Daten] [Shaoguo Liu, Haibo Wang, Jue Wang, Sunghyun Cho, Chunhong Pan] Existing image deblurring approaches often take the blur-kernel-size as an important manual parameter. When set improperly, this parameter can lead to significant errors in the estimated blur kernels. However, manually specifying a proper kernel size for an input image is usually a tedious trial-and-error process. In this paper, we propose a new approach for automatically estimating the underlying blur-kernel-size value that can lead to good kernel estimation. Our approach takes advantage of the autocorrelation map (automap) of image gradients that is known to reflect the motion blur information. We show that the standard automap suffers from structural edges in the image and cannot be directly used for kernel size estimation. To alleviate this problem, we develop a modified automap method that contains a directional attenuation component, which can effectively reduce the influence of structural edges, leading to more accurate and reliable kernel size estimation. Experimental results suggest that the proposed approach can help state-of-the-art deblurring algorithms achieve accurate kernel estimation without relying on manual parameter tweaking. Springer-Verlag Berlin Heidelberg, 2014 Image deblurring nationallicence Blur-kernel-size nationallicence Motion blur prior nationallicence Autocorrelation nationallicence Directional attenuation nationallicence Liu Shaoguo NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China aut Wang Haibo School of Control Science and Engineering, Shandong University, Jinan, China aut Wang Jue Adobe Research, Seattle, WA, USA aut Cho Sunghyun Adobe Research, Seattle, WA, USA aut Pan Chunhong NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China aut The Visual Computer Springer Berlin Heidelberg 31/5(2015-05-01), 733-746 0178-2789 31:5<733 2015 31 371 https://doi.org/10.1007/s00371-014-0998-2 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s00371-014-0998-2 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Liu Shaoguo NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China aut NATIONALLICENCE 700 1- Wang Haibo School of Control Science and Engineering, Shandong University, Jinan, China aut NATIONALLICENCE 700 1- Wang Jue Adobe Research, Seattle, WA, USA aut NATIONALLICENCE 700 1- Cho Sunghyun Adobe Research, Seattle, WA, USA aut NATIONALLICENCE 700 1- Pan Chunhong NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China aut NATIONALLICENCE 773 0- The Visual Computer Springer Berlin Heidelberg 31/5(2015-05-01), 733-746 0178-2789 31:5<733 2015 31 371